Elegant ways to rig multi-tap transformers?

[thermionic]

Hi,

I have an input transformer with 3 separate primaries. In order to get 3 different gain / impedance settings, there are combinations that can be used whereby you alternate between using a single tap, taps in series or taps in parallel.

Obviously, I can do this in a multitude of ways…

Conventional switches can suffer from stray capacitance, not to mention that you need contacts suitable for switching low-level signals (this is a high-gain application). Relays could do it, but that’s getting complicated as they’ll want a dedicated power rail. A Shallco would be great, but that’s uber-expensive for a stereo unit. Gold-plated jumpers will work well and they’re cheap – the downside will be that you’ll have to take off the lid to change tap and it’s not exactly studio-friendly if you intend to let other people use the device. I guess I could do it via an external tag strip?

What’s the most elegant way you’ve seen an OEM do it?

As per usual, I’m probably blind to an obvious, elegant solution…thanks in advance.



Justin

 

[bcarso]

What is the highest voltage it's going to see?

 

[thermionic]

Never more than a few millivolts.

Thanks,
J

 

[zebra50]

Hi Justin,

Not exactly what you asked but an old sony tape machine I saw had a cunning mains voltage selector which might be an inspiration.

It wired the transformer to a 6 pin socket. The plug was internally wired so that it could be removed, rotated and reinserted to select 110, 220 or 240V.

One might be able to use a number of plugs with differerent internal wiring to give different options. Drawback is losing the plugs!

 

[bcarso]

[quote author="thermionic"]Never more than a few millivolts.

Thanks,
J[/quote]

I once used 74HC4066 parts in a non-audio application to switch coils of an inductive pickup arrangement for narrowband magnetic field sensing. I found that if operated close to their maximum rails the on resistance of some manufacturers' parts was as low as 15 ohms or so. The capacitances are also quite low, and they can be mounted close to the things being switched to minimize strays.

The appeal here might be to operate the switches and associated logic from batteries, since the quiescent current is negligible. And you will not have the dry contact problems of mechanical or electromechanical switching.

 

[PRR]

> an input transformer... Conventional switches can suffer from stray capacitance

Bah. The self-capacitance of the winding, and probably the cabling, far exceeds switch capacitance.

Also switching on the primary of an input transformer is driven by the source, which usually assumes long cable to load and has significant drive ability.

Sure, you could screw-up the CMRR at 20KHz. But enough to matter in studio?

All non-solder techniques, even Shallco, will eventually have low-level issues. (You can clean a Shallco is the only advantage.) And jumpers will baffle users. I say pop in a $1.99 4P3T switch and see how bad it is. You may find that it is faultless for many years. The most likely failure is tarnish which will clear-up for a while when the frustrated user turns the knob back and forth several times, a soft-failure which will cue you to do something: blow another $1.99 or do something cleverer.

 

[thermionic]

Thanks. I will probably try a toggle, although I have had issues with them in low-level apps before. As PRR says, actuation backwards and forwards a few times tends to cure the issue, but - in the eyes of end-users - it doesn't inspire confidence as they tend to assume the switch is shot.

Justin